CN115829170A - Driving scheme optimization method, system and storage medium - Google Patents
Driving scheme optimization method, system and storage medium Download PDFInfo
- Publication number
- CN115829170A CN115829170A CN202310127795.1A CN202310127795A CN115829170A CN 115829170 A CN115829170 A CN 115829170A CN 202310127795 A CN202310127795 A CN 202310127795A CN 115829170 A CN115829170 A CN 115829170A
- Authority
- CN
- China
- Prior art keywords
- chromosomes
- chromosome
- target
- determining
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Navigation (AREA)
Abstract
The application provides a driving scheme optimization method, a driving scheme optimization system and a storage medium, wherein the driving scheme optimization method comprises the following steps: determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm. The technical scheme provided by the application can effectively combine the genetic algorithm with the big data information, and accurately and efficiently provide a relatively energy-saving driving scheme, so that the problem of the requirement on energy conservation and emission reduction of the commercial vehicle industry can be better solved.
Description
Technical Field
The present invention relates to an information processing method based on a specific calculation model, and more particularly, to a driving scheme optimization method, system, and storage medium.
Background
In recent years, the transportation industry in China is developed vigorously. In the transportation of people and goods by vehicles, passenger or freight system managers and passenger car or truck drivers often desire a driving scheme with the lowest energy consumption of vehicles on the transportation line under the condition of meeting the passenger or freight time limit requirements. In the aftermarket field of commercial vehicles, there have been proposals to improve the driving level by driving behavior evaluation guidance technology to improve the fuel economy of the vehicle. However, under the large background of energy saving and emission reduction, improving the economic driving effect of vehicles by only evaluating the driving behavior to improve the driving level has not met the demand of current social development. The market has higher demand for the technical scheme of the optimization of the new generation of driving scheme.
Disclosure of Invention
The application provides a driving scheme optimization method, which comprises the following steps: determining the vehicle speed interval of each travel of the target vehicle on the target route based on the operation data of the vehicle with the same type as the target vehicle, which is retrieved from the big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
According to the embodiment of the application, before determining the vehicle speed interval of each travel of the target vehicle on the target route, the driving scheme optimization method further includes: and dividing the target line into a plurality of sections of strokes based on the high-speed service area information and a preset mileage threshold value.
According to the embodiment of the application, dividing the target route into multiple sections of routes based on the high-speed service area information and the preset mileage threshold value comprises: dividing the target line into a plurality of sections of first strokes by using a high-speed service area as a segmentation point; and in response to any one of the plurality of sections of first strokes being larger than a preset mileage upper threshold limit, dividing the first stroke into a plurality of sections of second strokes.
According to an embodiment of the application, dividing the first stroke into a plurality of second strokes comprises: dividing the first stroke into a plurality of sections of sub-strokes according to the ascending slope, the flat road and the descending slope; sequentially judging whether the divided sub-strokes are smaller than a preset mileage threshold lower limit or not according to the stroke sequence, and sequentially combining the sub-strokes into subsequent sub-strokes in response to the divided sub-strokes being smaller than the preset mileage threshold lower limit until the combined sub-strokes are larger than or equal to the preset mileage threshold lower limit; and combining the last sub-stroke into the previous sub-stroke in response to the last sub-stroke being smaller than the preset mileage lower threshold.
According to the embodiment of the application, the step of determining the most energy-saving driving scheme for the target vehicle to complete the target route through a genetic algorithm comprises the following steps: randomly setting an initial vehicle speed for each section of travel on the target line in a vehicle speed interval of each section of travel; selecting a combination of initial vehicle speeds of all the sections of travel with the total time length of the target vehicle for completing the target route smaller than the longest time limit as a combination of a group of initial vehicle speeds in an initial vehicle speed setting set, wherein the initial vehicle speed setting set comprises M groups of combinations of the initial vehicle speeds; setting binary coding values of the initial vehicle speeds of all the strokes in each group of combination of the initial vehicle speeds in the initial vehicle speed setting set as all the gene values of all the segments of one chromosome, so as to obtain M first generation chromosomes; determining fitness of each chromosome based on the total energy consumption value corresponding to each chromosome in the M first generation chromosomes; sequentially selecting M chromosomes to be crossed from M chromosomes of a current generation by a roulette selection algorithm, wherein each chromosome in the M chromosomes can be repeatedly selected; performing random crossing operation on the M chromosomes to be crossed again to generate M crossed chromosomes; carrying out mutation operation on the M crossed chromosomes to generate M second generation chromosomes; performing the roulette selection operation, the crossover operation and the mutation operation on the M second generation chromosomes until M Nth generation chromosomes are generated; selecting a chromosome with the highest fitness from the first generation chromosomes to the Nth generation chromosomes; and decoding each segment of gene of the selected chromosome with the highest fitness into the target vehicle speed of each segment of travel.
According to an embodiment of the present application, randomly setting an initial vehicle speed for each trip on the target route within a vehicle speed interval of each trip comprises: and randomly setting an initial vehicle speed for each section of travel on the target line in the vehicle speed interval of each section of travel through a pseudo-random Sobol sequence.
According to an embodiment of the present application, determining the fitness of each chromosome based on the total energy consumption value corresponding to each chromosome of the M first generation chromosomes comprises: determining the total oil consumption Qi of the target vehicle for completing the target line based on each segment of genes of the ith chromosome in the M first generation chromosomes; according to
And determining the fitness Ai of the ith chromosome.
According to the embodiment of the application, sequentially selecting M chromosomes to be crossed from M chromosomes of the current generation by a roulette selection algorithm comprises the following steps of: determining a normalized probability of each chromosome being selected based on fitness of each chromosome in the M chromosomes of the current generation; selecting M chromosomes based on the normalized probability of each chromosome being selected; and sequentially listing the selected M chromosomes as the M chromosomes to be crossed, wherein each chromosome in the M chromosomes of the current generation can be repeatedly selected.
According to an embodiment of the present application, performing a random crossing operation on the M secondary chromosomes to be crossed, and generating M crossed chromosomes includes: pairwise matching the M chromosomes to be crossed in the selected sequence; determining whether to cross the paired chromosomes according to a self-adaptive cross probability function; in response to determining to cross the paired chromosomes, performing a random crossing operation on the paired chromosomes to form crossed chromosomes; determining the paired chromosomes as directly after the crossing in response to determining not to cross the paired chromosomesThe chromosome of (a); wherein the adaptive cross probability function P c Comprises the following steps:
P c1 and P c2 F' is the fitness of the more-fit of the two chromosomes in the pair, f is an empirical parameter max Is the fitness of the chromosome with the maximum fitness in the M chromosomes of the current generation, f avg The average fitness of the M chromosomes of the current generation.
According to an embodiment of the present application, performing mutation operations on the M crossed chromosomes to generate M second generation chromosomes includes: determining whether to mutate the chromosome according to a self-adaptive mutation probability function; generating a mutated chromosome by randomly flipping binary codes of one or more genes of the chromosome in response to determining to mutate the chromosome; in response to determining not to mutate the chromosome, directly determining the chromosome as the mutated chromosome; wherein the adaptive mutation probability function P m Comprises the following steps:
P m1 and P m2 F is the fitness of the chromosome to be mutated as an empirical parameter.
According to an embodiment of the present application, P c1 =0.85，P c2 =0.55，P m1 =0.1，P m2 =0.002。
According to the embodiment of the application, the driving scheme optimization method further comprises the following steps: after mutation operations are performed on the M crossed chromosomes of each generation, the K chromosomes with the highest fitness in the M mutated chromosomes of the current generation are subjected to gene optimization through a simulated annealing method to serve as initial chromosomes of the next round of roulette selection operation, crossing operation and mutation operation.
The application also provides a driving scheme optimization system, the driving scheme optimization system includes: a memory storing executable instructions; and one or more processors in communication with the memory to execute the executable instructions to: determining the vehicle speed interval of each travel of the target vehicle on the target route based on the operation data of the vehicle with the same type as the target vehicle, which is retrieved from the big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
The present application further provides a computer-readable storage medium for driving scenario optimization, wherein the computer-readable storage medium stores executable instructions that are executable by one or more processors to: determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
The technical scheme provided by the application can effectively combine the genetic algorithm with the big data information, and accurately and efficiently provide a relatively energy-saving driving scheme, so that the problem of the requirement on energy conservation and emission reduction of the commercial vehicle industry can be better solved.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic diagram of a driving scenario optimization method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a specific process for optimizing a driving scenario using a genetic algorithm according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a roulette selection algorithm according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a crossover algorithm according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a mutation algorithm according to an embodiment of the present application; and
fig. 6 is a schematic structural diagram of a driving scenario optimization system according to an embodiment of the present application.
Detailed description of the preferred embodiments
For a better understanding of the present application, the technical solutions thereof will be described in more detail with reference to the accompanying drawings. It should be understood that the detailed description is merely illustrative of exemplary embodiments of the present application and is not intended to limit the scope of the present application in any way. Like reference numerals refer to like elements throughout the specification. The expression "and/or" includes any and all combinations of one or more of the associated listed items.
In the drawings, the size, proportion and shape of the illustrations have been adjusted slightly for the convenience of illustration. The figures are purely diagrammatic and not drawn to scale. As used herein, the terms "approximately," "about," and the like are used as terms of table approximation and not as terms of table degree, and are intended to account for inherent deviations in measured or calculated values that will be recognized by those of ordinary skill in the art.
It will be further understood that expressions such as "comprising," "including," "having," "including," and/or "containing" are open-ended and not closed-ended expressions in this specification that indicate the presence of stated features, elements, and/or components, but do not preclude the presence or addition of one or more other features, elements, components, and/or groups thereof. Moreover, when a statement such as "at least one of" appears after a list of listed features, it modifies the entire list of features, rather than just a single feature in the list. Furthermore, when describing embodiments of the present application, the use of "may" mean "one or more embodiments of the present application. Also, the word "exemplary" is intended to refer to an example or illustration.
Unless otherwise defined, all terms (including engineering and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
In addition, the features of the embodiments and examples in the present application may be combined with each other without conflict. In addition, unless explicitly defined or contradicted by context, the specific steps included in the methods described herein are not necessarily limited to the order described, but can be performed in any order or in parallel. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1 is a schematic diagram of a driving scenario optimization method according to an embodiment of the present application.
Referring to fig. 1, the present application proposes a driving scenario optimization method 1000. According to the embodiment of the application, the driving scheme optimization method comprises the following steps.
In step S1010, a vehicle speed interval of each trip of the target vehicle on the target route is determined based on the operation data of the vehicle of the same type as the target vehicle retrieved from the big data cluster server. The vehicle operation data may be the same type of vehicle operation data including vehicle identification, vehicle operation time, longitude, latitude, direction, elevation, vehicle speed, rotational speed, torque, throttle opening, brake, load, etc. The vehicle operation data can be obtained from the big data cluster server through a big data extraction technology, and the big data extraction method is not limited in the application. The vehicle speed section of each section of journey can be vehicle speed section information calculated by the central server or the vehicle-mounted computer based on load information and power information of the target vehicle, gradient information of each section of journey of the target route and vehicle speed data of the same type of vehicle retrieved from the big data cluster server, and generally comprises a lower vehicle speed limit and an upper vehicle speed limit of the section of journey. According to the embodiment of the application, the vehicle speed interval of each section of travel can be determined after some interference data in the vehicle operation data are removed. The operation data of the vehicle of the same type as the target vehicle needs to be retrieved from the big data cluster server, so that the vehicle speed suggestion without reference value is set for the target vehicle in the driving scheme optimization process. For example, for a steep uphill road section or a mountain road section with many curves, if a vehicle speed advice is set for the target vehicle ignoring the road section situation, such a vehicle speed advice may exceed the traveling capability of the target vehicle. Therefore, referring to the operation data of the vehicle of the same type as the target vehicle, it is possible to avoid deviation of the finally set vehicle speed advice from the actual road condition.
In step S1020, a vehicle speed interval of each trip of the target vehicle on the target route and a maximum time limit for the target vehicle to complete the target route are used as constraint conditions, and a most energy-saving driving scheme for the target vehicle to complete the target route is determined through a genetic algorithm. The maximum time limit for the destination line is typically determined by the mission time limit requirements for passenger or freight transport. The method is characterized in that a genetic algorithm is adopted as a calculation model on the whole, the input quantity of the genetic algorithm can be the speed interval of each section of travel of the target vehicle on the target route, the longest time limit for the target vehicle to finish the target route, and the output quantity of the genetic algorithm is the most energy-saving speed value for the target vehicle to finish each section of travel of the target route. The genetic algorithm is an algorithm for simulating the global search optimal solution of the superior-inferior evolution rule of individuals in the natural population. It can seek an optimal solution through multiple iterations (i.e., "evolutions") under defined conditions. The technical scheme provided by the application can effectively combine the genetic algorithm with the big data information, and accurately and efficiently provide a relatively energy-saving driving scheme, so that the problem of the requirement on energy conservation and emission reduction of the commercial vehicle industry can be better solved.
As described above, in order to avoid that the finally set vehicle speed suggestion deviates from the actual road condition, the technical scheme provided by the application refers to the operation data of the vehicle with the same type as the target vehicle, so as to determine the vehicle speed interval of each section of travel of the target vehicle on the target route. In order to provide more accurate reference information and take account of calculation efficiency, the method and the device for dividing the target line into multiple sections of routes based on high-speed service area information and preset mileage threshold values are provided.
More specifically, according to the present application, the target link may be first divided into a plurality of trips in advance using the high speed service area as a preliminary segmentation point. Although the stroke division is convenient and quick, the accuracy is not good enough. This is because the distances between the high-speed service areas may have a large difference, the distances between some high-speed service areas may be only ten kilometers, and the distances between some high-speed service areas may reach hundreds of kilometers. Therefore, according to the improved technical scheme of the application, if any one of the first trips is larger than the preset mileage upper threshold limit (for example, 100 kilometers), the first trip is divided into multiple second trips. This division can be performed in the following manner.
Firstly, dividing the first stroke exceeding the preset mileage upper threshold into a plurality of sub-strokes according to the gradient. Such sub-trips may be, for example, an uphill road segment, a level road segment, a downhill road segment, a level road segment, an uphill road segment, in that order.
And then, sequentially judging whether the divided sub-strokes are smaller than a preset mileage threshold lower limit or not according to the stroke sequence. For example, if the first sub-trip is smaller than the preset mileage lower threshold, the sub-trip is merged with the next sub-trip. And if the combined sub-stroke is still smaller than the preset mileage lower threshold, continuing to combine the combined sub-stroke with the next sub-stroke until the combined sub-stroke is larger than or equal to the preset mileage lower threshold.
And according to the stroke sequence, continuing the sub-stroke combination process until each combined sub-stroke is greater than or equal to the preset mileage threshold lower limit.
However, in some partitioning processes, it is possible that the last sub-trip may eventually remain less than the preset lower mileage threshold since it has not had a subsequent sub-trip. In this case, it may be merged with the previous sub-stroke.
The difference of each section of travel in distance after being divided according to the rule is greatly reduced, so the referential of the target vehicle speed interval determined based on the operation data retrieved from the big data cluster server is also greatly improved.
Fig. 2 is a schematic diagram of a specific process for optimizing a driving scheme using a genetic algorithm according to an embodiment of the present application.
Referring to fig. 2, the present application proposes a genetic algorithm 2000 to optimize an automotive operating scheme. According to an embodiment of the present application, a genetic algorithm for optimizing an automotive operating scheme includes the following steps.
In step S2010, an initial vehicle speed is randomly set for each trip on the target route within the vehicle speed interval for each trip.
For example, the initial vehicle speed can be randomly set for each trip on the target line within the vehicle speed interval of each trip through the pseudorandom Sobol sequence. The pseudorandom Sobol sequence is a low-difference sequence, and the individual randomness is sacrificed to obtain the improvement of the individual uniformity. In the present application, a pseudo-random Sobol sequence is used to generate a random number Qn, where Qn ∈ [0,1]. Determining the initial vehicle speed Vn of each section of travel by using the random number Qn and combining the vehicle speed interval of each section of travel as follows:
Vn=Vmin+Qn×(Vmax-Vmin)。
in step S2020, a combination of initial vehicle speeds for respective trips in which the total length of time for the target vehicle to complete the target link is less than the longest time limit is selected as a combination of a set of initial vehicle speeds in an initialization vehicle speed setting set that includes combinations of M sets of initial vehicle speeds.
After the pseudo-random Sobol sequence is adopted to generate the random number Qn and the initial vehicle speed Vn, whether the combination of the initial vehicle speeds Vn can meet the total time length requirement of the target route or not needs to be checked, and if the combination of the initial vehicle speeds Vn meets the vehicle speed interval requirement of each section of journey, but the target vehicle cannot complete the target route within the longest time limit according to the combination of the initial vehicle speeds Vn, the combination of the initial vehicle speeds Vn is not satisfactory. Finally, from the combinations of the initial vehicle speeds Vn meeting the longest time limit requirement, M groups of combinations of the initial vehicle speeds can be selected as the basis for the "evolution" of the subsequent genetic algorithm. M is also referred to as the number of individuals in the population, or as the size of the population. According to the present application, M may be set to an even number between 50 and 70 in order to compromise accuracy and computational efficiency.
In step S2030, the binary code values of the initial vehicle speeds of the respective strokes in each set of combinations of the initial vehicle speeds in the set of the initial vehicle speed settings are set as the respective segment gene values of one chromosome, thereby obtaining M first-generation chromosomes.
The driving scheme can be digitalized through the binary coding, so that the driving scheme optimization method is converted into a mathematical problem for solving an optimal solution. Assuming that the maximum vehicle speed that the target vehicle can reach is 100km/h, binary encoding of the initial vehicle speed requires a 7-bit binary number. If the target line has a total of P strokes, the length of one chromosome is 7 XP. Through the transformation, the initial vehicle speed is transformed into a gene in biology, and the problem of solving the optimal vehicle speed combination can be transformed into the problem of solving the optimal chromosome by using a genetic algorithm.
In step S2040, fitness of each chromosome is determined based on the total energy consumption value corresponding to each chromosome in the M first generation chromosomes.
Fitness function is a function used in genetic algorithms to evaluate the "quality" of chromosomes (or "population individuals"). Higher fitness means that the chromosome quality is higher and the mathematical solution represented by the chromosome tends to be the optimal solution. In the optimization problem of the driving scheme, the total energy consumption value of the target vehicle for completing the target route is an important evaluation factor of the quality of the driving scheme. It is therefore a reasonable approach to determine the fitness function based on the total energy consumption value that the target vehicle needs to consume to complete the target route.
The following illustrates how the fitness function is determined using a fuel vehicle as an example.
The present application shows an exemplary calculation formula for the total oil consumption Q:
wherein: q is the total oil consumption value of the target vehicle for completing the target route, and the unit is L; q j Fuel consumption in liters (L) for a short distance (mathematically "minor") for the target vehicle to complete the target route; pj is the running power of the target vehicle during the short distance, and the unit is kW; ge is specific oil consumption, namely the weight of fuel oil required by unit energy consumption, and the unit is g/kWh; t is the time of the target vehicle for completing the short distance, and the unit is second(s); rho is the fuel density and is expressed in g/ml.
The calculation formula of Pj is as follows:
and Vj is the speed of the target vehicle at the time j and has the unit of km/h. F Drive the The function that is converted to the vehicle speed Vj can be approximated based on the parameter information of the target vehicle. Therefore, the total fuel consumption Q can be estimated based on the vehicle speed values of the respective trips.
The fitness of each chromosome may be determined according to the following formula:
wherein: ai is the fitness of the ith chromosome;Qithe total oil consumption of the target route is completed by the target vehicle according to the combination of the vehicle speeds determined by the ith chromosome.
After the fitness of each chromosome is determined, M chromosomes to be crossed next can be sequentially selected from M chromosomes of the current generation by a roulette selection algorithm at step S2050. Wherein each chromosome of the M chromosomes of the current generation can be selected repeatedly.
Referring to the wheel selection algorithm illustrated in fig. 3, the normalized probability of each chromosome being selected can be determined from the fitness Ai of each chromosome in the M chromosomes of the current generation. Fitness of chromosomesThe larger Ai, the larger the normalized probability and the larger the area occupied in the wheel. Then, a pointer is placed somewhere on the wheel. In this case, each time the wheel is rotated, the pointer points to a particular chromosome, e.g., E1, E2, ei, or E M . Each chromosome of the M chromosomes can be selected repeatedly. Thus, after M rotations of the wheel, M chromosomes to be crossed can be selected.
In step S2060, a random crossing operation is performed on the M chromosomes to be crossed next time, so as to generate M crossed chromosomes.
After selecting the M chromosomes to be crossed next, the chromosomes can be paired according to the selected order. For example, chromosome E1 selected for the first time is paired with chromosome E2 selected for the second time; pairing the third selected chromosome E3 with the fourth selected chromosome E4; chromosome E from selection M-1 M-1 With chromosome E selected at M time M And (6) pairing. Then, referring to fig. 4, one or more pairs of genes of the paired chromosomes are crossed. For example, the 1 st gene of chromosome E1 and the 1 st gene of chromosome E2 are interchanged with each other; for another example, the 1 st and 4 th genes of chromosome E1 are interchanged with the 1 st and 4 th genes of chromosome E2, respectively. The chromosomes E1 'and E2' obtained after the interchange are crossed.
According to another embodiment of the present application, the chromosome also needs to be validated after crossing. Specifically, after the intersection, it is necessary to determine whether the target vehicle can complete the target route within the maximum time limit according to the speed per hour of each trip corresponding to the chromosome after the intersection. If not, the corresponding paired chromosomes are re-randomly crossed until crossed chromosomes meeting the longest time limit are obtained.
According to another embodiment of the present application, before the paired chromosomes are crossed, it is determined whether the paired chromosomes are crossed. Specifically, whether or not to cross paired chromosomes can be determined according to an adaptive cross probability function. If it is finally determined that the paired chromosomes are crossed, performing a random crossing operation on the paired chromosomes as described above with reference to fig. 4 to form crossed chromosomes; and if finally determining that the paired chromosomes are not crossed, directly determining the paired chromosomes as crossed chromosomes.
The specific form of the adaptive cross probability function is as follows:
wherein, P c1 And P c2 As empirical parameters, e.g. P c1 =0.85，P c2 =0.55.f' is the fitness of the chromosome with the greater fitness of the two chromosomes in the pair, f max Is the fitness of the chromosome with the maximum fitness in the M chromosomes of the current generation, f avg Is the average fitness of the M chromosomes of the current generation.
In step S2070, mutation operations are performed on the M crossed chromosomes to generate M second generation chromosomes.
The mutation operation is an operation of rewriting the gene value of one or several genes of a chromosome. The operation is to break the restriction of the original gene combination, so that not only the local optimal solution can be found, but also the global optimal solution can be possibly found.
Referring to fig. 5, mutation operations may be performed on each of the M crossed chromosomes. During the mutation operation, one or more genes of the chromosome can be randomly selected for mutation. For example, in the example shown in fig. 5, mutation operation is performed on the first gene G1 of one chromosome after crossing. During the mutation operation, each gene value (i.e., binary code value of velocity) of the gene G1 is inverted between "0" and "1", thereby obtaining a mutated gene G1'. The other genes of the chromosome after the mutation are consistent with the genes before the mutation except the 1 st gene G1; the 1 st gene G1' of the chromosome after mutation has a gene value opposite to that of the original gene G1.
According to another embodiment of the present application, after mutation, the chromosome after mutation is required to be verified. Specifically, after the mutation, it is necessary to determine whether the target vehicle can complete the target route within the maximum time limit according to the time rate of each trip corresponding to the mutated chromosome. If not, the random mutation operation is carried out again on the corresponding chromosome until the mutated chromosome which can meet the longest time limit is obtained.
According to another embodiment of the present application, before performing the mutation operation, it is determined whether or not the mutation operation is performed on the chromosome. In particular, whether to mutate a chromosome can be determined according to an adaptive mutation probability function. If it is finally determined that the chromosome is mutated, randomly flipping binary codes of one or more genes of the chromosome by the method described with reference to fig. 5 to generate a mutated chromosome; and if finally determining that the chromosome is not mutated, directly determining the chromosome as the mutated chromosome.
The adaptive mutation probability function Pm is:
P m1 and P m2 As empirical parameters, e.g. P m1 =0.1，P m2 =0.002.f is the fitness of the chromosome to be mutated.
In step S2080, the roulette selection operation, the crossover operation, and the mutation operation are performed on the M second-generation chromosomes until M nth-generation chromosomes are generated. Through multi-generation evolution, the global optimal solution can be gradually approached. According to the application, in order to take account of accuracy and computational efficiency, N may be chosen to be a natural number between 400 and 500.
At step S2090, one chromosome with the highest fitness is selected from the first-generation chromosomes to the nth-generation chromosomes.
In each evolution, M evolved chromosomes (i.e., crossed and mutated chromosomes) are obtained. After N generations of evolution, M × N chromosomes can be obtained. Therefore, the chromosome with the highest fitness can be selected from the M × N chromosomes.
In step S2100, each segment gene of the selected chromosome having the highest fitness is decoded as the target vehicle speed for each segment trip. Due to the negative correlation between the fitness and the energy consumption, the speed combination scheme that the target vehicle completes the target route with the lowest energy consumption can be obtained after each section of gene of the chromosome with the highest fitness is decoded into the target speed of each section of travel.
According to another embodiment of the present application, after mutation operations are performed on M crossed chromosomes of each generation, the K chromosomes with the highest fitness among the M mutated chromosomes of the current generation are genetically optimized by a simulated annealing method to serve as initial chromosomes for the next round of roulette selection, crossing and mutation operations.
The simulated annealing method is a random optimization algorithm based on a Monte-Carlo iterative solution strategy, is a probabilistic solution algorithm, and has the advantages of high calculation efficiency and strong local search capability. The algorithm is added in the evolution of the genetic algorithm, so that the efficiency and the quality of the genetic algorithm can be effectively improved, and the iteration times of the genetic algorithm are reduced. According to the application, parameters such as the starting temperature Tb, the ending temperature Tc, the cooling speed Vr and the like in the simulated annealing algorithm can be adjusted to perform gene optimization on the K chromosomes with the highest fitness in each generation of chromosomes.
The application also provides a driving scheme optimization system which can be realized in the forms of a mobile terminal, a Personal Computer (PC), a tablet computer, a server and the like. Referring now to FIG. 6, a schematic diagram of a drive schedule optimization system suitable for use in implementing embodiments of the present application is shown.
As shown in fig. 6, the computer system includes one or more processors, communication sections, and the like, for example: one or more Central Processing Units (CPUs) 601, and/or one or more image processors (GPUs) 613, etc., which may perform various appropriate actions and processes according to executable instructions stored in a Read Only Memory (ROM) 602 or loaded from a storage section 608 into a Random Access Memory (RAM) 603. Communications portion 612 may include, but is not limited to, a network card, which may include, but is not limited to, an IB (Infiniband) network card.
The processor may communicate with the ROM 602 and/or RAM 603 to execute the executable instructions, connect with the communication part 612 through the bus 604, and communicate with other target devices through the communication part 612, so as to complete the operations corresponding to any one of the methods proposed by the embodiments of the present application, for example: determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the longest time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
In addition, in the RAM 603, various programs and data necessary for the operation of the device can also be stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. The ROM 602 is an optional module in the presence of the RAM 603. The RAM 603 stores executable instructions, or writes executable instructions into the ROM 602 during running, and the executable instructions cause the CPU 601 to execute operations corresponding to the driving scheme optimization method. An input/output interface (I/O interface) 605 is also connected to bus 604. The communication unit 612 may be integrated, or may be provided with a plurality of sub-modules (e.g., a plurality of IB network cards) and connected to the bus link.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage unit 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary.
It should be noted that the architecture shown in fig. 6 is only an optional implementation manner, and in a specific practical process, the number and types of the components in fig. 6 may be selected, deleted, added or replaced according to actual needs; in different functional component settings, separate settings or integrated settings may also be used, for example, the GPU and the CPU may be separately set or the GPU may be integrated on the CPU, the communication section 612 may be separately set or integrated on the CPU or the GPU, and so on. These alternative embodiments are all within the scope of the present disclosure.
In particular, according to the present application, the process described with reference to fig. 1 may be implemented as a computer program product. For example, the present application proposes a computer program product comprising computer readable instructions which, when executed by a processor, implement the following: determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server; and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
In such embodiments, the computer program product may be downloaded and installed from a network via the communication section 609 and/or read and installed from the removable medium 611. The computer program product performs the above-mentioned functions defined in the method of the present application when executed by the CPU 601.
The solution of the present application may be implemented in many ways. For example, the technical solution of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The order of the steps used to describe the method is provided for clarity of description of the embodiments only. Unless specifically limited, the method steps of the present application are not limited to the order specifically described above. Furthermore, in some embodiments, the present application may also be implemented as a storage medium storing a computer program product.
The above description is only an embodiment of the present application and an illustration of the technical principles applied. It will be appreciated by a person skilled in the art that the scope of protection covered by the present application is not limited to the embodiments with a specific combination of the features described above, but also covers other embodiments with any combination of the features described above or their equivalents without departing from the technical idea described above. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (14)
1. A driving scheme optimization method is characterized by comprising the following steps:
determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server;
and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
2. The driving scheme optimization method according to claim 1, further comprising, before determining the vehicle speed intervals for each trip of the target vehicle on the target line:
and dividing the target line into a plurality of sections of strokes based on the high-speed service area information and a preset mileage threshold value.
3. The driving scenario optimization method of claim 2, wherein dividing the target route into a plurality of trips based on high speed service area information and a preset mileage threshold comprises:
dividing the target line into a plurality of sections of first strokes by using a high-speed service area as a segmentation point;
and in response to any one of the plurality of sections of first strokes being larger than a preset mileage upper threshold limit, dividing the first stroke into a plurality of sections of second strokes.
4. The method of claim 3, wherein dividing the first stroke into the plurality of second strokes comprises:
dividing the first stroke into a plurality of sections of sub-strokes according to the ascending slope, the flat road and the descending slope;
judging whether the divided sub-strokes are smaller than a preset mileage threshold lower limit or not in sequence according to the stroke sequence, and combining the sub-strokes into subsequent sub-strokes in sequence in response to the divided sub-strokes being smaller than the preset mileage threshold lower limit until the combined sub-strokes are larger than or equal to the preset mileage threshold lower limit;
and combining the last sub-stroke into the previous sub-stroke in response to the last sub-stroke being smaller than the preset mileage lower threshold.
5. The driving scheme optimization method according to claim 1, wherein determining the most energy-saving driving scheme for the target vehicle to complete the target route through a genetic algorithm comprises:
randomly setting initial vehicle speed for each section of travel in the vehicle speed interval of each section of travel on the target line;
selecting a combination of initial vehicle speeds of all the sections of travel with the total time length of the target vehicle for completing the target route smaller than the longest time limit as a combination of a group of initial vehicle speeds in an initial vehicle speed setting set, wherein the initial vehicle speed setting set comprises M groups of combinations of the initial vehicle speeds;
setting binary coding values of the initial vehicle speeds of all the strokes in each group of combination of the initial vehicle speeds in the initial vehicle speed setting set as all the gene values of all the segments of one chromosome, so as to obtain M first generation chromosomes;
determining fitness of each chromosome based on the total energy consumption value corresponding to each chromosome in the M first generation chromosomes;
sequentially selecting M chromosomes to be crossed from M chromosomes of a current generation by a roulette selection algorithm, wherein each chromosome in the M chromosomes can be repeatedly selected;
performing random crossing operation on the M chromosomes to be crossed again to generate M crossed chromosomes;
carrying out mutation operation on the M crossed chromosomes to generate M second generation chromosomes;
carrying out the roulette selection operation, the crossover operation and the mutation operation on the M second-generation chromosomes until M Nth-generation chromosomes are generated;
selecting a chromosome with the highest fitness from the first generation chromosomes to the Nth generation chromosomes;
and decoding each segment of gene of the selected chromosome with the highest fitness into the target vehicle speed of each segment of travel.
6. The driving scenario optimization method according to claim 5, wherein randomly setting an initial vehicle speed for each trip on the target route within the vehicle speed interval of each trip comprises:
and randomly setting an initial vehicle speed for each section of travel on the target line in the vehicle speed interval of each section of travel through a pseudo-random Sobol sequence.
7. The driving scenario optimization method of claim 5, wherein determining the fitness of each chromosome based on the total energy consumption value corresponding to each chromosome of the M first generation chromosomes comprises:
determining the total oil consumption Qi of the target vehicle for completing the target line based on each segment of genes of the ith chromosome in the M first generation chromosomes;
according to the following
And determining the fitness Ai of the ith chromosome.
8. The driving scenario optimization method of claim 5, wherein sequentially selecting M chromosomes to be crossed from M chromosomes of a current generation by a roulette selection algorithm comprises:
determining a normalized probability of each chromosome being selected based on fitness of each chromosome in the M chromosomes of the current generation;
selecting M chromosomes based on the normalized probability of each chromosome being selected;
and sequentially listing the selected M chromosomes as the M chromosomes to be crossed, wherein each chromosome in the M chromosomes of the current generation can be repeatedly selected.
9. The driving scenario optimization method of claim 8, wherein the randomly crossing the M chromosomes to be crossed next time to generate M crossed chromosomes comprises:
pairwise matching the M chromosomes to be crossed in the selected sequence;
determining whether to cross the paired chromosomes according to a self-adaptive cross probability function;
in response to determining to cross the paired chromosomes, performing a random crossing operation on the paired chromosomes to form crossed chromosomes;
in response to determining not to cross the paired chromosomes, directly determining the paired chromosomes as the crossed chromosomes;
wherein the adaptive cross probability function P c Comprises the following steps:
P c1 and P c2 F' is the fitness of the more-fit of the two chromosomes in the pair, f is an empirical parameter max Is the fitness of the chromosome with the maximum fitness in the M chromosomes of the current generation, f avg Average of M chromosomes for the current generationAnd (4) fitness.
10. The driving scheme optimization method according to claim 9, wherein performing mutation operations on the M crossed chromosomes to generate M second generation chromosomes comprises:
determining whether to mutate the chromosome according to a self-adaptive mutation probability function;
generating a mutated chromosome by randomly flipping binary codes of one or more genes of the chromosome in response to determining to mutate the chromosome;
in response to determining not to mutate the chromosome, directly determining the chromosome as the mutated chromosome;
wherein the adaptive mutation probability function P m Comprises the following steps:
P m1 and P m2 F is the fitness of the chromosome to be mutated as an empirical parameter.
11. The method of claim 10, wherein P is c1 =0.85，P c2 =0.55，P m1 =0.1，P m2 =0.002。
12. The driving scenario optimization method according to claim 5, further comprising:
after mutation operations are carried out on the M crossed chromosomes of each generation, the K chromosomes with the highest fitness in the M mutated chromosomes of the current generation are subjected to gene optimization through a simulated annealing method to serve as initial chromosomes of the next round of roulette selection operation, crossing operation and mutation operation.
13. A driving scenario optimization system, comprising:
a memory storing executable instructions; and
one or more processors in communication with the memory to execute the executable instructions to:
determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server;
and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the maximum time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
14. A computer-readable storage medium for driving scenario optimization, the computer-readable storage medium storing executable instructions that are executable by one or more processors to:
determining a vehicle speed interval of each travel of a target vehicle on a target route based on operation data of vehicles with the same types as the target vehicle, which are retrieved from a big data cluster server;
and determining the most energy-saving driving scheme for the target vehicle to finish the target route by using the speed interval of each section of travel of the target vehicle on the target route and the longest time limit for the target vehicle to finish the target route as constraint conditions through a genetic algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310127795.1A CN115829170B (en) | 2023-02-17 | 2023-02-17 | Driving scheme optimization method, driving scheme optimization system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310127795.1A CN115829170B (en) | 2023-02-17 | 2023-02-17 | Driving scheme optimization method, driving scheme optimization system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115829170A true CN115829170A (en) | 2023-03-21 |
| CN115829170B CN115829170B (en) | 2023-05-12 |
Family
ID=85521744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310127795.1A Active CN115829170B (en) | 2023-02-17 | 2023-02-17 | Driving scheme optimization method, driving scheme optimization system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115829170B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120197504A1 (en) * | 2010-12-23 | 2012-08-02 | Cummins Intellectual Property, Inc. | System and method of vehicle speed-based operational cost optimization |
| CN106503804A (en) * | 2016-10-11 | 2017-03-15 | 南京理工大学 | A kind of train timing energy-saving operation method based on Pareto multi-objective genetic algorithms |
| CN110893853A (en) * | 2018-08-23 | 2020-03-20 | 厦门雅迅网络股份有限公司 | Vehicle control method and system based on front gradient information |
| CN111145068A (en) * | 2019-12-09 | 2020-05-12 | 南京航空航天大学 | Long-distance high-timeliness economical cruise vehicle speed planning method |
| CN111191813A (en) * | 2019-10-14 | 2020-05-22 | 重庆邮电大学 | Vehicle distribution path optimization method based on cargo load and soft time window limitation |
| US20210366287A1 (en) * | 2020-05-20 | 2021-11-25 | Nature Mobility Co., Ltd. | System for providing big data-based artificial intelligence automatic allocation matching service using taxi demand prediction |
| CN113805570A (en) * | 2020-05-28 | 2021-12-17 | 广州汽车集团股份有限公司 | Collaborative planning method and system for vehicle running path and running speed and storage medium |
| CN114444843A (en) * | 2021-11-30 | 2022-05-06 | 上海海洋大学 | Agricultural product green logistics distribution vehicle scheduling method and system based on large-scale variable neighborhood search strategy |
-
2023
- 2023-02-17 CN CN202310127795.1A patent/CN115829170B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120197504A1 (en) * | 2010-12-23 | 2012-08-02 | Cummins Intellectual Property, Inc. | System and method of vehicle speed-based operational cost optimization |
| CN106503804A (en) * | 2016-10-11 | 2017-03-15 | 南京理工大学 | A kind of train timing energy-saving operation method based on Pareto multi-objective genetic algorithms |
| CN110893853A (en) * | 2018-08-23 | 2020-03-20 | 厦门雅迅网络股份有限公司 | Vehicle control method and system based on front gradient information |
| CN111191813A (en) * | 2019-10-14 | 2020-05-22 | 重庆邮电大学 | Vehicle distribution path optimization method based on cargo load and soft time window limitation |
| CN111145068A (en) * | 2019-12-09 | 2020-05-12 | 南京航空航天大学 | Long-distance high-timeliness economical cruise vehicle speed planning method |
| US20210366287A1 (en) * | 2020-05-20 | 2021-11-25 | Nature Mobility Co., Ltd. | System for providing big data-based artificial intelligence automatic allocation matching service using taxi demand prediction |
| CN113805570A (en) * | 2020-05-28 | 2021-12-17 | 广州汽车集团股份有限公司 | Collaborative planning method and system for vehicle running path and running speed and storage medium |
| CN114444843A (en) * | 2021-11-30 | 2022-05-06 | 上海海洋大学 | Agricultural product green logistics distribution vehicle scheduling method and system based on large-scale variable neighborhood search strategy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115829170B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shao et al. | Electric vehicle‐routing problem with charging demands and energy consumption | |
| TWI638320B (en) | Systems, methods and non-transitory computer-readable storage mediums for recommending an estimated time of arrival | |
| CN109416878B (en) | System and method for recommending estimated time of arrival | |
| Letnik et al. | Dynamic management of loading bays for energy efficient urban freight deliveries | |
| CN110520913B (en) | System and method for determining estimated time of arrival | |
| CN105160428B (en) | The planing method of electric automobile on highway quick charge station | |
| Tu et al. | OCD: Online crowdsourced delivery for on-demand food | |
| Jung et al. | Design and modeling of real-time shared-taxi dispatch algorithms | |
| TW201741993A (en) | System and method for determining routes of transportation service | |
| Ma et al. | A model for the stop planning and timetables of customized buses | |
| CN110059875B (en) | Public bicycle demand prediction method based on distributed whale optimization algorithm | |
| Zhan et al. | A modified artificial bee colony algorithm for the dynamic ride-hailing sharing problem | |
| CN111340318B (en) | Vehicle dynamic scheduling method and device and terminal equipment | |
| CN112381472B (en) | Subway connection bus route optimization method and device and storage medium | |
| CN111126799B (en) | Shared network driver and crew matching method based on bipartite graph | |
| CN111191899B (en) | Vehicle scheduling method based on region division parallel genetic algorithm | |
| CN112183812A (en) | Finished cigarette logistics vehicle scheduling method considering short-time and low-cost | |
| CN111932341A (en) | Method and system for determining car pooling order | |
| Tang et al. | Energy-optimal routing for electric vehicles using deep reinforcement learning with transformer | |
| CN115829170B (en) | Driving scheme optimization method, driving scheme optimization system and storage medium | |
| CN116562581A (en) | Rail transit connection bus route and scheduling optimization method based on shared bicycle travel influence | |
| CN116797126A (en) | Rural terminal logistics site selection-path planning method based on double-layer planning | |
| Rangriz et al. | A novel approach to optimize the ridesharing problem using genetic algorithm | |
| Yu et al. | Comparison of advantages and disadvantages among various algorithms in logistics path design—Taking H-group as an example | |
| Guo et al. | How to find a comfortable bus route-Towards personalized information recommendation services |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |